Sediment Dynamics and Lead (Pb) Contamination in an Urban Estuarine River System

2.1 Study area

The study was conducted between June and August 2024 at two primary research locations: the Banjir Kanal Timur estuary and the Lamper River (see Figure 1 and Figure 2). Sampling in the estuarine area was conducted at three observation stations: Station E1 (6°57'6.81''S, 110°26'25.48''E), Station E2 (6°57'6.63''S, 110°26'25.06''E), and Station E3 (6°57'6.44''S, 110°26'24.60''E). In addition, one sampling point was established in the midstream section of the Lamper River, namely Station L (7°0'6.48"S, 110°26'39.20"E). Discharge measurements were conducted at the Banjir Kanal Timur Bridge, located approximately 5.06 km upstream from the estuary at 6°59'46.17"S and 110°26'20.91"E. This location was selected because of its hydrological stability, as it is not influenced by tidal fluctuations, thereby ensuring more reliable data. The site also offers practical advantages, including ease of access and a relatively shallow river profile, which facilitates more efficient sampling by minimizing constraints related to depth variability and tidal interference.

11.png

2.png

Figure 2. Flowchart of the data collection method

The Banjir Kanal Timur River primarily functions as a receiving body for runoff generated from various anthropogenic activities, including industrial, residential, and commercial sources. These characteristics formed the basis for its selection as the main study site. The river is highly suitable for investigating sediment transport and contaminant dynamics because of its significant sediment accumulation and high susceptibility to heavy-metal pollution. The Lamper River, as a tributary of the Banjir Kanal Timur, was also included to assess upstream flow contributions and compare pollution levels between interconnected river segments. Together, these two locations provide an integrated study area that enables a comprehensive analysis of sediment transport processes and pollution distribution along different sections of the river system in Indonesia.

2.2 Data collection method

A total of 12 independent sediment samples were collected over three months, from June to August 2024. Sampling was conducted at four points each month: three estuarine stations (E1, E2, and E3) and one upstream station in the Lamper River (L). Sediment samples were classified into two main categories: bed load and suspended load. Bed-load sampling followed ISO 5667-12:2017, while suspended-load sampling followed the procedures specified in ISO 5667-17:2000. All samples were immediately stored in acid-washed polyethylene containers, preserved at 4 ℃ in a cool box, and transported to the laboratory within 12 hours for further analysis.

For granulometric analysis, dried samples (105 ℃ for 24 hours) were first separated using a sieve shaker with mesh sizes ranging from 4.75 mm to 0.063 mm to obtain the sand fraction. The finer fraction (<0.063 mm), containing silt and clay, was further analyzed using the pipette method based on Stokes’ law, following ISO 1127:2020 (E). Gravimetric analysis was used to quantify sediment mass with reference to SNI 06-69398-2004. River discharge measurements were conducted based on SNI 03-2414-1991. Discharge was measured at a single representative point located approximately 5.06 km upstream of the estuary, at the Banjir Kanal Timur Bridge, which is not influenced by tidal backwater. For estuarine stations, discharge was assumed to be equal to this upstream value because no major tributaries occur between the measurement point and the sampling locations. However, the calculated transport rates are presented as estimates based on representative discharge rather than station-specific measured loads. Acid digestion was conducted according to SNI 8910:2021 to ensure optimal extraction of heavy metals from the samples. The concentration of Pb in the sediments was then determined using procedures outlined in SNI 06-6992.3-2004.

2.3 Laboratory analysis of sediment samples

Sediment samples were collected and classified into two primary groups: bed load and suspended load. Bed-load sampling was conducted in accordance with ISO 5667-12:2017, while suspended-load sampling followed the procedures outlined in ISO 5667-17:2000. The collected samples were subsequently analyzed through two main stages: granulometric analysis and gravimetric analysis. Granulometric analysis was performed to determine particle-size distribution, following ISO 1127:2020 (E). The percentages of sand, silt, and clay obtained from the granulometric analysis were subsequently used to classify sediment texture using the soil texture triangle, as illustrated in Figure 3. Gravimetric analysis was used to quantify sediment mass with reference to SNI 06-69398-2004. River discharge measurements were conducted based on SNI 03-2414-1991, which provides guidance for measuring discharge in open channels. Acid digestion was conducted according to SNI 8910:2021 to ensure optimal extraction of heavy metals from the samples. Heavy-metal concentrations in the sediments were then determined using the procedures outlined in SNI 06-6992.3-2004 to evaluate the level of heavy-metal contamination in sediment media.

3.png

Figure 3. Classification of sediment types using a triangular diagram

Source: Modified from research [19].

2.4 Sediment transport calculation

Sediment transport analysis in this study focused on both suspended load and bed load using established empirical formulations [20, 21]. The suspended sediment transport rate was calculated using Eq. (1), which incorporates key hydraulic and sediment parameters, including flow velocity, sediment density, median grain size, and bed shear stress. This formulation reflects the balance between hydrodynamic forces that promote sediment movement and gravitational forces that resist particle suspension.

$q_{s w}=0.05 \times \gamma_s \times v^2 \times\left[\frac{D_{50}}{g\left(\frac{\gamma_s}{y}-1\right)}\right]^{\frac{1}{2}} \times\left[\frac{\tau_0}{\left(\gamma_s-y\right) D_{50}}\right]^{\frac{3}{2}}$           (1)

In this equation, $q_{s w}$ represents the suspended sediment transport rate (ton/day), which is controlled primarily by flow velocity ($v$), sediment density ($\gamma_s$), and median diameter of sediment particles $\left(D_{50}\right)$. The parameter $\gamma$ denotes water density, while $g$ represents gravitational acceleration. Bed shear stress $\left(\tau_0\right)$, which governs sediment entrainment, was calculated using Eq. (2):

$\tau_0=\gamma \times h \times S$          (2)

Here, bed shear stress $\left(\tau_0\right)$ is influenced by water density $(\gamma)$, flow depth $(h)$, and channel slope $(S)$, indicating that deeper flows and steeper gradients enhance sediment mobilization.

Bed-load transport was estimated using Eq. (3), which relates sediment movement along the riverbed to discharge and channel characteristics:

$q_b=\left(39.5 \times Q^{\frac{2}{3}} \times S \times 9.95 \times d\right)^{\frac{3}{2}}$              (3)

In this formulation, $q_b$ denotes the bed-load transport rate (ton/day), which is primarily governed by river discharge (Q), channel slope (S), and representative sediment grain size (d). Higher discharge increases the capacity of the flow to transport sediment along the bed.

The empirical formulas used in this study were selected because they are widely applied to lowland river systems due to their simplicity and minimal data requirements. However, these formulations were originally developed for non-tidal, unidirectional flow conditions. Therefore, their application in estuarine environments, where tidal influences may affect flow direction and velocity, requires careful interpretation. The results presented in this study should therefore be regarded as estimates rather than absolute measured values because they were not verified against direct field measurements of sediment transport.

River discharge measurements were conducted to support sediment transport calculations by determining flow velocity and cross-sectional characteristics during each sampling event. Flow velocity was measured using a flow meter, while discharge was calculated as the product of flow velocity and wetted cross-sectional area, following SNI 03-2414-1991 [22]. The wetted cross-sectional area was obtained by dividing the river into multiple segments to account for spatial variability in depth and width, and by summing the section areas of all segments.

Flow velocity at each vertical was determined using standard point-measurement techniques depending on flow depth, ensuring a representative velocity distribution across the cross-section. This segmented approach allows for a more accurate estimation of discharge by capturing variations in flow conditions within the river channel. Overall, the integration of field measurements and empirical formulations provides a reliable basis for evaluating sediment transport dynamics under natural flow conditions.

2.5 Heavy metal analysis

Heavy-metal analysis in this study was conducted using a combination of SNI standards and the EPA 2002 method. The procedure began with wet digestion using an acid mixture to extract metals from sediment samples. Approximately 2-5 g of a representative sample was placed in a digestion vessel, followed by the addition of 3 mL nitric acid (HNO₃, 65%) and 9 mL hydrochloric acid (HCl, 37%). The mixture was then heated at 120 ℃ to ensure complete decomposition of the sediment matrix and release bound metals into solution. After cooling, the digested sample was filtered using 2.5 μm filter paper to remove residual particulates. The filtrate was subsequently transferred into a 50 mL volumetric flask and diluted with deionized water to achieve a consistent final volume.

Pb concentrations in the samples were determined using Atomic Absorption Spectrophotometry (AAS) with a Buck Scientific 210 VGP instrument (USA). Background correction and matrix modification were applied to improve analytical accuracy. A blank solution containing the same acid composition as the digested samples, namely HCl and HNO₃ diluted to 50 mL, was used to stabilize the baseline. Absorbance measurements were performed at wavelengths of 283.3 nm and 217 nm, with final calibration conducted at 217 nm. To prevent cross-contamination, the aspirator was cleaned between sample measurements. A calibration curve was established using standard solutions, ensuring a correlation coefficient greater than 0.995 to maintain analytical reliability.

Quality assurance and quality control (QA/QC) procedures were applied to ensure analytical reliability. The method detection limit (MDL) and method quantification limit (MQL) for Pb were determined at 0.05 mg/kg and 0.15 mg/kg, respectively. Certified reference material (CRM), namely NIST SRM 2702 Marine Sediment, was analyzed with each batch to verify accuracy, with recoveries ranging from 92% to 107%. All samples were analyzed in duplicate, and the relative standard deviation (RSD) was maintained below 5%. Blank solutions and calibration standards were analyzed every 10 samples to correct for instrument drift. Iron (Fe) was measured independently for each sediment sample using the same AAS method to enable Enrichment Factor (EF) calculations.

The level of sediment contamination was evaluated using the I_geo, which assesses the degree of heavy-metal accumulation relative to background concentrations [23]. In this study, a background value of 0.4 mg/kg was used as the reference [24]. The I_geovalue was calculated using Eq. (4):

$I_{g e o}=\log _2\left[\frac{C_x}{1.5 C_h}\right]$             (4)

In this equation, $I_{g e o}$ represents the Geoaccumulation Index, which indicates the level of sediment contamination. C_x refers to the measured concentration of heavy metals in sediment (mg/kg), whileC_b denotes the natural background concentration (mg/kg). The constant factor of 1.5 is introduced to account for natural variations in background values due to lithogenic effects. Based on the calculated I_geo values, contamination levels are classified into categories ranging from uncontaminated to extremely contaminated.

The level of sediment contamination at each sampling location was evaluated using the I_geo, with classifications based on the criteria presented in Table 1 [25, 26].

Table 1. Classification of Geoaccumulation Index (I_geo) values [27]

Anthropogenic contributions to heavy-metal concentrations in sediments were further assessed using the EF [27]. EF quantifies the degree of metal enrichment relative to natural background levels by normalizing the concentration of the target element against a conservative reference element. In this study, iron (Fe) was used as the reference element, with a measured concentration of 26.308 mg/kg and a background value of 56.3 mg/kg. EF was calculated using Eq. (5) [28]:

$E F=\frac{\left(\frac{C_i}{C_{\text {ref }}}\right) \text { sample }}{\left(\frac{B_i}{B_{\text {ref }}}\right) \text { backgkround }}$              (5)

In this equation, EF represents the Enrichment Factor, indicating the extent of anthropogenic influence. $C_i$ refers to the concentration of the target metal in the sample, while $C_{\text {ref }}$ denotes the concentration of the reference element (Fe) in the sample. Similarly, $B_i$ and $B_{\text {ref }}$ represent the background concentrations of the target metal and the reference element, respectively.

EF values were classified into seven categories, as summarized in Table 2.

Table 2. Classification of Enrichment Factor (EF) value [29]

In addition, the C_f  was used to evaluate the degree of contamination associated with specific heavy metals [29]. A background value of 20 mg/kg was applied in this analysis. C_f was calculated using Eq. (6):

$C_f=\frac{C_i}{B_i}$                (6)

Here, $C_f$ represents the Contamination Factor, which reflects the level of contamination relative to background conditions. $C_i$ denotes the measured concentration of the heavy metal in sediment (mg/kg), while $B_i$ corresponds to its background concentration (mg/kg).

4.1 Sediment transport rates

Sediment transport rates exhibited distinct spatial and temporal variations, influenced by a combination of physical and hydraulic factors during the observation period. Key controlling parameters include flow velocity, channel slope, sediment particle-size distribution, channel geometry, and tidal forcing. Flow velocity plays a fundamental role in determining the capacity of a river to transport sediment, particularly coarse materials such as sand. Steeper channel slopes increase gravitational energy, thereby enhancing the capacity of the flow to mobilize and transport sediment particles [45, 46]. In addition, D₅₀ serves as an indicator of sediment stability, with coarser particles tending to resist movement and finer particles being more easily entrained and transported in suspension [45]. The interaction among these factors governs sediment transport dynamics across different locations.

Spatial variations were evident among the observation stations. Station E3 consistently exhibited relatively higher bed-load transport rates, which may be associated with nearby concrete structures that potentially induce local flow turbulence and promote sediment resuspension [47, 48]. In contrast, Estuary 1 showed relatively low sediment mobility, which may be influenced by local hydraulic conditions and dense riparian vegetation that enhances sediment retention. Vegetation along riverbanks likely contributes to flow resistance, stabilizes the banks, and reduces flow velocity near the margins, thereby enhancing sediment deposition [49]. However, field measurements of bed roughness and near-bank velocity were not conducted; therefore, these explanations should be considered qualitative inferences. This finding is consistent with previous studies [45], which highlight the significant role of riparian vegetation in regulating sediment flux and stabilizing fluvial systems.

Tidal dynamics further increase the complexity of sediment transport processes in estuarine environments. The study area experiences a diurnal tidal regime, resulting in periodic changes in flow direction and velocity. During flood-tide conditions, reduced flow velocity promotes sediment deposition, whereas during ebb flow, increased flow velocity facilitates downstream sediment transport [50]. This pattern was particularly evident at observation stations 2 and 3. In contrast, the Lamper station, located outside the tidal influence zone, exhibited relatively stable and lower sediment transport rates due to unidirectional flow conditions.

These findings are consistent with previous studies conducted in urban river systems [51], which demonstrate that anthropogenic modifications, such as impermeable surfaces and engineered structures, significantly influence sediment transport processes. Seasonal variability also plays an important role, as highlighted by Patriadi et al. [52], who reported that sediment transport is affected by changes in hydrological conditions and geomorphological characteristics, including slope and vegetation distribution. In this study, dry-season conditions likely enhanced the relative influence of channel morphology and sediment sources, resulting in observable variations in sediment transport across the study area.

4.2 Heavy metal concentration

The concentration of heavy metals in sediment samples in this study is represented by lead (Pb) levels. Based on Table 7, Pb concentrations measured in the Banjir Kanal Timur River and the Lamper River were generally higher than those reported for several other locations. Although these concentrations remain below the TEL, further evaluation using the I_geo, EF, and C_f indicates that anthropogenic activities continue to influence sediment quality. The highest Pb concentration was recorded at Estuary 3 in August, suggesting localized pollutant accumulation. This elevated concentration may be associated with increased land-based activities during the dry season, when reduced river discharge limits the capacity of the water body to dilute contaminants, thereby promoting pollutant accumulation over time.

Sediment characteristics also play a significant role in controlling Pb distribution. Fine-grained particles, particularly silt and clay, have a greater capacity to adsorb heavy metals, resulting in higher Pb concentrations under conditions of elevated TSS [53]. In addition, tidal currents influence the transport and redistribution of suspended particles, thereby affecting Pb distribution patterns within the estuarine system [54]. Local infrastructure, including drainage networks and embankments, further contributes to pollutant dynamics by altering runoff pathways and facilitating the mixing of point and non-point pollution sources.

A notable discrepancy was observed between high $I_{g e o}$ values (4.65-5.19) and moderate $C_f$ values (0.755-1.096). This discrepancy is primarily attributable to the different background values applied. $I_{\text {geo }}$ uses a conservative local baseline background of $0.4 \, \mathrm{mg} / \mathrm{kg}$ [24, 55], employs a logarithmic scale, and incorporates a correction factor of 1.5 to account for natural lithogenic variations, thereby yielding elevated values. In contrast, $C_f$ is calculated as a simple ratio against the global average baseline of $20 \, \mathrm{mg} / \mathrm{kg}$ reported by Wedepohl [56], producing more conservative estimates. Consequently, $I_{\text {geo}}$ indicates "heavily to extremely contaminated" conditions, whereas $C_f$ suggests only low to moderate contamination.

Table 6. Comparison of Pb concentrations from the Semarang coastal area and surrounding locations

The primary sources of Pb contamination in the study area are associated with industrial activities and urban development, particularly automotive-related operations, informal metal processing, and small-scale manufacturing. Additional contributions may arise from vehicle brake wear, tire residues, and the recycling of batteries and electronic waste. Previous studies have shown that metal pollution in similar environments often originates from multiple sources, including natural, traffic-related, industrial, and mixed contributions [58]. Isotopic analyses further indicate that Pb contamination may be dominated by natural sources and coal combustion, with a smaller contribution from traffic emissions [59].

A comparison of Pb concentrations with other locations in the Semarang coastal region (see Table 7) shows that Pb levels in the Lamper River and estuarine stations (E1–E3) were generally higher than those reported in areas such as the Trimulyo Channel, Mangunharjo Channel, and the eastern coast of Semarang [56, 57]. These values also exceed concentrations observed in the Banjir Kanal Barat estuary [55], although they remain slightly lower than the maximum values reported in previous studies of the Banjir Kanal Timur [53]. Furthermore, Pb concentrations in sediments were significantly higher than those reported in biota, such as Perna viridis from Tanjung Mas Harbor and surrounding waterways [54]. Despite these differences, all measured concentrations remain below the sediment quality guideline threshold of 35 mg/kg.

The apparent discrepancy between high I_geo values (4.652–5.190) and relatively low to moderate C_f values (0.755–1.096) can be attributed to differences in calculation approaches and reference baselines. The I_geo index employs a logarithmic scale and incorporates a correction factor, typically 1.5, to account for natural geochemical variability, which tends to produce higher values. In addition, the choice of background concentration significantly influences the results. This study uses a relatively low baseline value of 0.4 mg/kg [58], compared with the global average of 20 mg/kg [59], resulting in elevated I_geo values. In contrast, C_f is calculated as a simple ratio between measured and background concentrations, producing more conservative estimates. Consequently, I_geo indicates “heavily to extremely contaminated” conditions, whereas C_f suggests only moderate contamination.

The EF values obtained in this study (1.61–2.34) indicate minor to moderate enrichment, suggesting a measurable but not dominant contribution from anthropogenic sources. Because EF values are independent of absolute background values, they confirm that anthropogenic inputs are present but relatively limited. The highest EF value was observed at Estuary 3, which is located near industrial and automotive-related activities, supporting the interpretation of localized human influence. In contrast, lower EF values at Estuary 1 may reflect the mitigating effect of riparian vegetation and reduced pollutant input from upstream areas. Therefore, the combined interpretation of all three indices suggests that the study area is best characterized as experiencing low to moderate Pb contamination with localized anthropogenic pressure, rather than the “extremely contaminated” category implied by I_geo alone.

Potential sources of Pb contamination in the study area may be associated with surrounding land-use activities, including automotive workshops, small-scale metal industries, and domestic runoff. However, because no source-apportionment analysis, such as isotope analysis, detailed land-use inventory, or discharge survey, was conducted in this study, these attributions should be considered possible sources based on typical urban pollution patterns rather than definitive conclusions.

Overall, Pb concentrations in the Banjir Kanal Timur and Lamper River are consistent with other reports from Java, although they are lower than concentrations reported in heavily polluted sites such as Tirang Beach (49.16 mg/kg) and the Citarum River, where concentrations reach up to 80 mg/kg. Conversely, significantly lower Pb levels have been reported in less impacted areas such as the Madura Strait (0.12–2.01 mg/kg), reflecting minimal anthropogenic influence [58]. Based on the combined interpretation of C_f, I_geo, and EF values, the study area can be classified as experiencing moderate pollution with notable anthropogenic pressure.

Urban water bodies in Semarang tend to exhibit elevated Pb concentrations due to intensive land use and urban runoff, consistent with patterns observed in other Indonesian urban systems. In contrast, offshore environments typically show lower Pb concentrations due to dilution by seawater, sediment deposition in low-energy zones, and geochemical processes such as oxidation and complexation [60]. Tidal processes also contribute to the redistribution of metals across wider areas, leading to localized decreases in concentration, as observed in Jakarta Bay [61]. A similar mechanism is likely present in the Banjir Kanal Timur estuary, where sediments act as temporary sinks for heavy metals before their eventual transport to marine environments.